Variants of the Genetic Algorithm on Load Frequency Control Application

A Genetic Algorithm (GA) is a soft computing technique influenced by biological processes. The core concept of GA consists of two fundamental elements: individuals and populations. In this context, “population” refers to a group of individual solutions that need to be optimized. Building on this foundation, this chapter introduces three distinct variations of the Genetic Algorithm (GA) in the selection procedure: roulette wheel selection, tournament selection, and uniform crossover selection. These proposed variations are thoroughly evaluated within the context of load frequency control problems in single-area, dual-area, and three-area thermal power systems. The outcomes of simulations are carefully scrutinized to determine which variation performs most effectively for specific problem types. Later, the top-performing variation is chosen and integrated with well-established algorithms, including Artificial Bee Colony (ABC) and Cohort Intelligence (CIO). The selection of the most suitable variation holds the potential to not only strengthen the performance of these algorithms but also to broaden the application of soft computing techniques in addressing various real-world problems.